whisper-medium-paper

hts98/whisper-medium-paper

whisper-medium-paper is a automatic speech recognition model from hts98. Use it when you need speech turned into text. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

166

Public

Repo size

30.6 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin6.1 GB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
Model type: whisper
Access: Public
Created: Jun 17, 2023
Updated: Jun 19, 2023
SHA: 4655691b

Model card

From the Hugging Face model README

whisper-medium-paper

This model is a fine-tuned version of hts98/whisper-medium-paper on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.4741
Wer: 38.0826

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 13
eval_batch_size: 13
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	1.0	264	0.3606	39.1061
0.1341	2.0	528	0.3903	38.9291
0.1341	3.0	792	0.4337	39.0249
0.0469	4.0	1056	0.4426	38.3089
0.0469	5.0	1320	0.4741	38.0826

Framework versions

Transformers 4.28.0.dev0
Pytorch 2.0.0+cu117
Datasets 2.10.1
Tokenizers 0.13.2

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
Created: Jun 17, 2023
Updated: Jun 19, 2023